Nucleic acids, proteins and polysaccharides are tree major classes of biopolymers. While the first two systems are principally linear assemblies, polysaccharides are structurally more complex. This structural and stereochemical diversity results in a rich content of "information" in relatively small molecules. Nature further "leverages" the structural versatility of polysaccharides by their covalent attachment (i.e. "conjugation") to other biomolecules such as isoprenoids, fatty acids, neutral lipids, peptides or proteins.
Oligosaccharides in the form of glycoconjugates mediate a variety of events including inflammation, immunological response, metastasis and fertilization. Cell surface carbohydrates act as biological markers for various tumors and as binding sites for other substances including sub pathogens.
More specifically, an increasing number of physiologically important recognition phenomena involving carbohydrates have been discovered in recent years. Lectins, proteins which contain carbohydrate recognition domains, have been identified. Prominent members of the calcium dependent (C-type) lectin family (Drickamer, K. Curr. Opin. Struct. Biol. 1993, 3, 393) are the selectins which play a crucial role in leukocyte recruitment in inflammation. Bevilacqua, M. P.; Nelson, R. M. J Clin. Invest. 1993, 91, 379. Members of the C-type lectin superfamily have been described on NK cells and Ly-49, NKR-P1 and NKG2 constitute group V of C-type lectins. While many lectins have been purified and cloned, their ligands have not been identified due to the heterogeneous nature of carbohydrates.
The increasing recognition of the key roles of oligosaccharides and glycoconjugates in fundamental life sustaining processes has stimulated a need for access to usable quantities of these materials. Glycoconjugates are difficult to isolate in homogeneous form from living cells since they exist as microheterogeneous mixtures. The purification of these compounds, even when possible, is at best tedious and is generally achieved in very small yields. Given the travails associated with isolation from natural sources, a major opportunity for chemical synthesis presents itself.
Currently three powerful glycosylating agents are commonly used in the synthesis of oligosaccharides in solution and on the solid support. Trichloroacetimidates have been used for over fifteen years for the synthesis of oligosaccharides in solution and very recently on the solid support. The drawback of these excellent synthons is their lengthy synthesis.
Thioethyl glycosides have also been used successfully for the synthesis of olihgosaccharides in solution and on the solid support. Drawbacks are the use of the toxic stench ethanethiol during the synthesis and the requirement for methyl triflate, a carcinogen, as an activator. These drawbacks make the commercialization of this otherwise very attractive class of glycosylating agents nearly impossible.
Glycosyl sulfoxides involve the use of toxic thiols during their synthesis but otherwise can be activated by non-toxic agents. Both the synthesis of oligosaccharides in solution as well as on the solid support has been accomplished using this approach.
The invention of solid phase peptide synthesis by Merrifield 35 years ago dramatically influenced the strategy for the synthesis of biopolymers. The preparation of structurally defined oligopeptides (Atherton, E.; Sheppard, R. C. Solid phase peptide synthesis: A practical approach; IRL Press at Oxford University Press: Oxford, England, 1989, pp 203) and oligonucleotides (Caruthers, M. H. Science 1985, 230, 281) has benefited greatly from the feasibility of conducting their assembly on various polymer supports. The advantages of solid matrix based synthesis, in terms of allowing for an excess of reagents to be used and in their facilitation of purification are now well appreciated. It is obvious, that the level of complexity associated with the synthesis of an oligosaccharide on a polymer support dwarfs that associated with the other two classes of repeating biooligomers. First, the need to differentiate similar functionality (hydroxyl or amino) in oligosaccharide construction is much more challenging than is the situation with oligopeptides or oligonucleotides. Furthermore, in these latter two cases, there is no stereoselection associated with construction of the repeating amide or phosphate bonds. To the contrary, each glycosidic bond to be fashioned in a growing oligosaccharide ensemble constitutes a new locus of stereogenicity.
Remarkably, a great deal of progress had been achieved in assembling relatively complex carbohydrate ensembles on a solid support. Advances along these lines have involved the need for considerable simplification and refinement of protecting group strategies and the development of glycosylation methodology which is workably stereoselective and amenable to being conducted with one component anchored to an insoluble matrix.
The development of protocols for the solid support synthesis of oligosaccharides and glycopeptides requires solutions to several problems. Of course, considerable thought must be addressed to the nature of the support material. The availability of methods for attachment of the carbohydrate from either the "reducing" or "non-reducing" end would be advantageous. Also, selection of a linker which is stable during the synthesis, but can be easily cleaved when appropriate, is critical. A protecting group strategy that allows for high flexibility is desirable. Most important is the matter of stereospecific and high yielding coupling reactions.
Combinatorial chemistry has been used in the synthesis of large numbers of structurally distinct molecules in a time and resource efficient manner. Peptide, oligonucleotide, and small molecule libraries have been prepared and screened against receptors or enzymes to identify high-affinity ligands or potent inhibitors. For a review see: Thompson, L. A.; Ellman, J. A. Chem. Rev. 1996, 96, 555.
Generation of biologically active oligosaccharide libraries presents several interesting challenges. Each glycosidic bond to be fashioned in a growing oligosaccharide constitutes a new locus of stereogenicity, unlike the joining of nucleosides and peptides. Furthermore, the natural mammalian sugar monomers (FIG. 3) all carry at least three hydroxyl groups which can undergo glycosylation. Extensive branching, sulfation and phosphorylation of oligosaccharides are common in nature.
Two different strategies for the generation of combinatorial oligosaccharide libraries have been reported to date. The first approach followed the "random-glycosylation" strategy which is based on the assumption that all hydroxyls of an glycosyl acceptor react at the same rate. Kanie, O.; Barresi, F.; Ding, Y.; Labbe, J.; Otter, A.; Forsberg, L. S.; Ernst, B.; Hindsgaul, O. Angew. Chem. Int. Ed. Engl. 1995, 34, 2720. While "random-glycosylation" requires only a very limited number of monosaccharide building blocks, the analysis of the resulting mixtures poses an almost insurmountable problem. For the "site-specific" glycosylation approach to combinatorial oligosaccharide synthesis, differentially protected monosaccharides are employed. In this manner, only one particular hydroxyl group on the monosaccharide may be exposed and coupled. Either, each member of the library is synthesized in a separate reaction vessel (spatially separate synthesis method) or pooling strategies are employed to generate large libraries of compounds (split synthesis method). Very recently a libraxy of approximately 1500 modified oligosaccharides was prepared by the split synthesis method. Liang, R.; Yau, L.; Loebach, J.; Ge, M.; Uozumi, Y.; Sekanina, K.; Horan, N.; Gildersleeve, J.; Thompson, C.; Smith, A.; Biswas, K.; Still, W. C.; Kahne, D. Science 1996, 274, 1520. The library was screened for ligands of a lectin, while the compounds were still attached to the solid support. A tagging system was used to rapidly determine the structure of the selected compounds.
The ability to generate diverse combinatorial libraries containing carbohydrates is directly linked to the ability to prepare complex carbohydrates and therefore to the availability of potent glycosylation reactions. Currently three powerful glycosylating agents are commonly used in the synthesis of oligosaccharides in solution and on the solid support. Trichloroacetimidates have been used for over fifteen years for the synthesis of oligosaccharides in solution and very recently on the solid support. The drawback of these excellent synthons is their lengthy synthesis.
Thioethyl glycosides have also been used successfully for the synthesis of oligosaccharides in solution and on the solid support. Drawbacks are the use of the toxic stench ethanethiol during the synthesis and the requirement for methyl triflate, a carcinogen, as an activator. These drawbacks make the commercialization of this otherwise very attractive class of glycosylating agents nearly impossible.
Glycosyl sulfoxides involve the use of toxic thiols during their synthesis but otherwise can be activated by non-toxic agents. Both the synthesis of oligosaccharides in solution as well as on the solid support has been accomplished using this approach.
Thioethyl glycosides and sulfoxides have been used in the synthesis of oligosaccharide libraries. In both cases relatively small diaccharide (.about.1,500 compounds) and trisaccharide (.about.50 compounds) libraries were generated. Wong, C.-H.; Ye, X.-S.; Zhang, Z. J. Am. Chem. Soc. 1998, 120, 7137. The effort to synthesize the building blocks restricts the amount and the variety of starting materials that can be produced. The possibility to fashion monosacchaide building blocks in an efficient and straightforward fashion from glycal precursors presents a dramatic advantage over existing methods.
The random-glycosylation method by Hindsgaul et al. does not require differentially protected building blocks but does produce mixtures which make screening and identification of library composition impossible. This method is expected to have very limited practical use.
Several challenges have to be met to prepare combinatorial carbohydrate libraries. Synthetic strategies in which either the glycosyl donor or the glycosyl acceptor is attached to the solid support will be employed. A wide variety of differentially protected monosaccharide building blocks have to be prepared. Efficient glycosylation reactions have to be employed. The resulting libraries can be screened for lectin binding while still on the solid support or after already being cleaved.
The generation of combinatorial carbohydrate libraries will facilitate the rapid identification of ligands to many carbohydrate binding proteins which are involved in a variety of important biological events including inflammation (Giannis, A. Angew. Chem. Int. Ed. Engl. 1994, 33, 178), immune response (Ryan, C. A. Proc. Natl. Acad. Sci. U.S.A. 1994, 91, 1) and metastasis (Feizi, T. Curr. Opin. Struct. Biol. 1993, 3, 701). Analogs of ligands can help to define important lectin-ligand interactions. Non-natural ligands can be powerful inhibitors of carbohydrate-protein binding and will facilitate the study of cascade-like events involving such interactions. Furthermore, inhibitors of carbohydrate-lectin binding are potential candidates for a variety of therapeutic applications.